Study On Multicriteria Method For Water Resources Problems

Multicriteria decision making is of important in the content of decision theory and modern decision science. The exploitation, utilization, restoration, protection and management of water resources must involve many evaluation and decision making problems. Therefore studying the multicriteria decision analysis of water resources management systematically is of significance for reducing damage of flood and drought and improving benefits. Based on optimal multicriteria decision theory, model and method, this paper mainly focuses on the multi-criteria optimal decision of water resources and environment system. The major contents and research results are as follows:1. The multicriteria automatic parameter calibration model of the Xinanjiang model with single procedure is proposed. This method integrates two parts of Xinanjiang rainfall-runoff model calibration together and automatically calibrates and validates the parameters, taking the peak discharge, peak time and total runoff volume as the calibration criteria and using TOPSIS (Technique for Order Performance by Similarity to Ideal Solution) for GAs' fitness function. Using this model to calibrate Xinanjiang model can directly assess the model performance, calibrate the model parameter automatically, and significantly improve the efficiency of calibration. Taking Shuangpai Reservoir as an example to validate this method, it approves that this model is an effective and efficient method.2. This paper establishes a hybrid model which combines Dynamic Programming and MCDEA (Multicriteria Data Envelopment Analysis) model together for the multicriteria multireservoirs operations. The operation of reservoirs takes into account benefit and flood control criteria. The model is established regarding to a typical flood process. Dynamic programming model is used to calculate the state variables of the upstream reservoir and relevant downstream reservoir. At first the weights of criteria are determined, and then the MCDEA model is used to rank and evaluate the combined alternatives of up and down stream reservoirs. Taking Fengman-Baishan reservoirs as an example, the feasible alternative is presented. It is indicated that the new model can efficiently combine the dynamic programming and MCDEA together. It is an effective approach to solve the multicriteria multistage problem.3. The new Multiple Criteria Data Envelopment Analysis (MCDEA) model is established to solve the environment impact assessment (EIA) problems, which combines the quantitative and qualitative criteria together. The EIA usually assess quantitative factors, but the dam location selection consists of many qualitative and quantitative factors. In this paper aqualitative method is applied to transform the qualitative performance of alternatives to relative membership degree of qualitative factors. Then MCDEA model is used to rank the alternatives by considering the relative membership degree of qualitative factors as one of the quantitative data. A case study of dam location selection indicates the effectiveness of the proposed methodology. It is a simple and important method of EIA.4. TOPSIS (Technique for Order Performance by Similarity to Ideal Solution) model under the intuitionistic set is established using standard Hamming distance and standard Euclidean distance. It is applied to the alternatives selection problem of water supply project. In this paper, every alternative, every criterion, and every criteria weight are expressed by intuitionistic fuzzy number. Two distance calculating methods are applied to calculate the alternative index of decision model. Then the best alternative is determined. Comparing to the result of AHP, this model can describe the fuzzy characteristics of the engineering problem exquisitely, and model the practice precisely. It is an efficient method to solve the project comparing problem.5. The water resource allocation problem is a multicriteria decision problem. There are various uncertainties existing in the decision making problem, which include the uncertainties of criteria weights and criteria performance values. Sensitivity analysis can be used to analyse the impacts of these uncertainties on the ranking of alternatives. This paper conducts a sensitivity analysis for the muticriteria decision making parameters such as the weights and the criteria performance values. A reliability analysis is performed by randomly sampling of simulating parameters. A significance analysis is taken by the Spearman rank correlation coefficient to determine the extent of satisfaction of the relative best alternative. The benefits of the approach are illustrated by applying it to a case study.Finally, a summary is given and some problems to be further studied are discussed.